Robot Floor Plan Mapping Using Overlapping Depth Views
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mapping methods for robotic devices, such as EKF-based SLAM, face computational challenges, require high processing power, memory, and are costly, and traditional sensor-based methods limit navigation to perimeter mapping, lacking efficiency for consumer-grade robots.
Innovation Solution
A computationally efficient method using depth perception from cameras or depth sensors to construct floor plans by aligning and combining overlapping fields of view, reducing computational costs and enabling full-environment mapping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If EKF-based SLAM methods are used to construct environmental maps, then mapping completeness and accuracy are improved, but computational power requirements and memory usage increase significantly
Solution Approach 1:
The patent extracts only the essential depth information needed for mapping from the sensor data, rather than processing all feature points. By using depth sensors to capture distance measurements directly, the system extracts the critical mapping data while discarding unnecessary visual features, thereby reducing computational load while maintaining mapping accuracy.
Solution Approach 2:
The patent employs cost-effective depth sensors instead of expensive sophisticated SLAM systems. By using simpler, more affordable sensors that provide direct depth measurements, the system achieves acceptable mapping accuracy without the high computational requirements and costs of EKF-based methods, making the technology viable for consumer-grade robots.
2Measurement precision
If EKF-based SLAM methods are used to construct environmental maps, then mapping completeness is improved, but processing speed and robot task performance decrease due to computational delays
Solution Approach 1:
The system extracts only depth measurements from the environment rather than processing complete images with numerous feature points. This extraction approach captures the essential spatial information needed for mapping while dramatically reducing the data volume requiring processing, thereby improving robot task performance without sacrificing mapping completeness.
Solution Approach 2:
The patent replaces the computationally intensive EKF-based SLAM system with a simpler depth-sensing approach. By substituting complex mathematical filtering with direct depth measurement and basic spatial reasoning, the system achieves comparable mapping completeness while eliminating computational delays that hinder robot productivity.
3Measurement precision
If traditional sensor-based mapping methods are used, then mapping capability is achieved, but robot navigation is limited to perimeter mapping only
Solution Approach 1:
The patent transitions from two-dimensional perimeter tracking to three-dimensional depth-based mapping. By utilizing depth sensors that capture distance measurements in three dimensions, the system can construct comprehensive environmental maps that enable navigation throughout the entire workspace, not just along perimeters, thereby improving navigation flexibility and adaptability.
4Measurement precision
If sophisticated mapping techniques are used, then mapping accuracy is improved, but implementation cost increases significantly
Solution Approach 1:
The patent employs inexpensive depth sensors and simple processing algorithms instead of costly sophisticated SLAM systems. By using affordable components that provide sufficient mapping accuracy for consumer applications, the system achieves cost-effective implementation while maintaining acceptable mapping precision, making autonomous robots accessible to mass markets.
Data Source
AI summary
Provided is a medium storing instructions that when executed by a processor of a robot effectuate operations including: capturing, with a camera, a plurality of images of a working environment; obtaining image data of a first image and image data of a second image; determining an overlapping area of a field of view of the first image and a field of view of the second image by comparing the image data of the first image and the image data of the second image; spatially aligning the image data of the first image and the image data of the second image based on the overlapping area to construct a floor plan; inferring a geometry of the working environment based on image data corresponding with the plurality of images used in constructing the floor plan; and determining a path of the robot based on the floor plan.